Development and application of in vitro models of hematopoiesis to drug development.

In vitro models of hematopoiesis are used increasingly in investigative hematopathology. Such models complement in vivo animal testing and have been shown to be predictive for hematotoxicity associated with anticancer and antiviral agents in humans. In vitro models of hematopoiesis consist of short-term cloning assays for various hematopoietic progenitor cells and long-term functional assays for the marrow hematopoietic microenvironment. In our laboratories, the cloning assays have been used as investigative tools to study the pathogenetic mechanisms of drug-induced blood disorders and as screening systems to investigate the possible hematotoxic potential of candidate drugs in various animal species. Data in support of these applications are presented in this paper.

General pharmacology of loracarbef in animals.

Loracarbef ((6R, 7S)-7-[(R)-2-amino-2-phenyl-acetamido]-3-chloro-8-oxo-1- azabicyclo [4.2.0]oct-2-ene-2-carboxylic acid, monohydrate, LY 163892, CAS 121961-22-6) is a carbacephem antibiotic targeted for use in the treatment of infectious disease. The potential pharmacological effects of this agent were examined on cardiovascular, respiratory, gastrointestinal, central nervous and autonomic nervous systems. Also examined were local anesthetic activity, effects on platelet aggregation, circulating blood glucose, primary antibody production, renal function, blood coagulation, ocular irritation, and the acute inflammatory response. Doses of 100, 1000, and 2000 mg/kg given by the oral route were selected for most in vivo studies. Concentrations up to 3 x 10(-3) mol/l were used in vitro. Loracarbef was essentially inactive in the tests of central and autonomic nervous system function, platelet aggregation, renal function, blood hemolysis, primary antibody production, blood coagulation, and ocular irritation. It had no local anesthetic activity. At high oral or intravenous doses, representing significant multiples of the therapeutic dose, loracarbef caused changes in gastrointestinal (decrease in gastric acid production and gastric fluid volume; increased biliary output), cardiovascular (increased mean pressure, cardiac output, heart rate, and femoral flow), blood glucose (increased glucose levels), and anti-inflammatory tests (suppressed acute inflammatory response). In summary, loracarbef exhibited minimal activity in these pharmacodynamic studies. These results indicate loracarbef has a low potential to produce adverse effects at therapeutic doses.

General pharmacology of pergolide in animals. 2nd communication: gastrointestinal, renal and miscellaneous studies.

Pergolide mesylate ((8 beta)-8-[(methylthio)methyl]-6-propylergoline monomethanesulfonate, LY 127809, CAS 66104-23-2) is a novel and potent dopamine agonist marketed for treating the symptoms of Parkinson's disease. The potential secondary pharmacological effects of this agent on the gastrointestinal and renal systems, as well as effects on local anesthesia, hemolysis, platelet aggregation, circulating blood glucose, primary antibody production, and the acute inflammatory response were examined. Pergolide exhibited significant pharmacological effects in gastrointestinal, renal and anti-inflammatory tests at high oral doses. Pergolide was essentially inactive in blood hemolysis, platelet aggregation, primary antibody production and local anesthesia testing. In summary, these studies confirm the pharmacological selectivity of pergolide, and indicate a low potential for secondary pharmacological side effects upon the functions tested at clinically relevant doses.

Canine BFU-e progenitors: adaptation of a reproducible assay and anatomical distribution.

In vitro cloning assays are used increasingly in investigative hematotoxicology and in screening candidate compounds for their hematotoxic potential. To expand these applications, a practical cloning assay for erythroid burst-forming units (BFU-e) that uses a microplasma clot (MPC) system was adapted to the dog, a species used extensively in experimental hematology and drug development. This system offers the advantage over the methylcellulose and soft agar culture systems of allowing specimen fixation and, therefore, morphological and cytochemical evaluation. The distribution of BFU-e among various anatomic sites was assessed using the MPC cloning system, which was modified to optimize the BFU-e growth. BFU-e growth required only erythropoietin (Epo) in the culture medium and there was no need for an exogenous source of burst-promoting activity (BPA). The cloning efficiency was linearly proportional to the plating concentrations of Epo and marrow mononuclear cells (MMC) over a range of 0 to 3 U Epo and 1 x 10(5) to 3 x 10(5) MMC per ml of culture, respectively. Increases in concentrations of Epo and MMC beyond these levels were not associated with linear growth. The addition of transferrin and spleen-conditioned medium containing a mixture of growth factors (including BPA) reduced BFU-e growth. The relative concentration of BFU-e was comparable among samples collected from the iliac crest, femur, and humerus. Serial cultures performed on individual dogs were highly reproducible and there was little variation in BFU-e activity among dogs of comparable age. It was concluded that the MPC system is a practical and reproducible cloning system for early (BFU-e), as well as late erythroid colony-forming units (CFU-e) in the dog. The concentration of BFU-e appears comparable throughout the active marrow; therefore, various anatomic sites can be used interchangeably for serial quantitative analysis of this progenitor.

Comparative distribution of marrow CFU-e and CFU-gm progenitors in different anatomic sites in the dog.

The interpretation of marrow cloning activity, particularly in serial cultures, is greatly influenced by the reproducibility of the collected marrow samples. In order to establish whether bone marrow cloning activities and precision of the cloning assays are influenced by the site of bone marrow collection in the dog, we studied the incidence of marrow erythroid (CFU-e) and granulocyte-macrophage (CFU-gm) progenitor cells in the iliac crest, sternum, vertebrae, femur, and humerus, using microplasma clot and soft agar culture systems. Marrow samples obtained from the femur and humerus revealed consistently higher cell concentrations than those from the iliac crest, vertebrae, or sternum. Those aspirated from the sternum and vertebrae had lower cell concentrations and were less reproducible. Statistical analysis revealed no significant differences in the incidence of marrow CFU-e and CFU-gm progenitor cells between the femur, humerus, iliac crest or vertebrae. With multiple sampling, the marrow cloning efficiency was consistent and reproducible within the individual dogs. We conclude that the distribution of CFU-e and CFU-gm is comparable throughout the active marrow in the dog and that these sites may be used interchangeably for multiple quantitative analysis of marrow hematopoietic progenitor cells.

General pharmacology of nizatidine in animals.

Nizatidine (N-[2-[[[2-[(dimethylamino)methyl-4-thiazolyl]- methyl]thio]ethyl]-N'-methyl-2-nitro-1,1-ethenediamine, LY139037, Axid) is a novel, potent, and selective H2-antagonist. The potential secondary pharmacologic effects of this agent on the cardiovascular, respiratory, gastrointestinal, renal, hepatic, autonomic, and central nervous systems as well as effects on circulating blood glucose and the acute inflammatory response were examined. Nizatidine was generally inactive in the tests conducted in mice, rats, guinea pigs, rabbits, and dogs. Nizatidine and the reference H2-antagonist, cimetidine, both produced effects upon the cardiovascular and respiratory systems by intravenous administration in anesthetized dogs at doses in excess of the intended clinical exposure. In summary, these studies confirm the selective pharmacologic activity of nizatidine and indicate a low potential for secondary pharmacologic side effects to be encountered clinically.

Electron microscopic study of the unique features and structural-morphologic relationship of canine bone marrow.

Electron microscopy revealed several unique features in canine bone marrow, compared with that of other species. The marrow was fatty and extensively trabeculated and was enclosed by a complete layer of endosteal bone-lining cells. Branched reticular cells were closely associated with each other and, occasionally, covered part of the sinus wall as an adventitial layer. The extent of adventitial coverage varied markedly and was less extensive, compared with that of other species. On average, only 23% of the sinus wall was covered by adventitial layer, in contrast to 65% reported in laboratory animals. Unilaminar sinuses, with no adventitial coverage, accounted for greater than 38% of all sinuses. Quantitative analysis indicated that 60% of the latter sinuses contained apertures, as opposed to 35% of sinuses with adventitial coverage (P less than 0.05). Moreover, the number of apertures in unilaminar sinuses was significantly (P less than 0.009) greater than that in multilaminar sinuses. Apertures were observed every 59 microns in unilaminar sinuses, in contrast to every 109 microns in multilaminar sinuses. Approximately 75% of the apertures were occupied by cells in transit, and only 25% were free of cells. Macrophages were distributed throughout the marrow and were closely associated with all blood cell lines. Occasionally, cells that entered the lumen were not fully mature. Erythroblasts were seen migrating across the wall and within the lumen of sinuses. The less-extensive adventitial coverage in canine bone marrow might indicate that the rate of cell delivery from the marrow into the circulation was relatively high in this species.(ABSTRACT TRUNCATED AT 250 WORDS)

Cephalosporin-induced alterations in erythroid (CFU-E) and granulocyte-macrophage (CFU-GM) colony-forming capacity in canine bone marrow.

Cephalosporins are among the safest antibiotics. Nevertheless, hematologic abnormalities ranging from mono- to pancytopenia do occur, albeit infrequently, following their therapeutic use. Similar abnormalities to those reported in people have been seen in dogs given high doses of cephalosporins. As part of a study to define the latter thoroughly, we explored the effects of long-term, high-dose cephalosporin administration on canine marrow erythroid (CFU-E) and granulocyte-macrophage (CFU-GM) progenitor cells. Cefazedone (Refosporen, E. Merck, Darmstadt) was administered intravenously at doses of 540 to 840 mg/kg daily to 14 healthy beagle dogs for up to 4 months, or less if hematologic effects were evident earlier. Within 6 to 10 weeks, treated dogs developed pancytopenia (5/14), thrombocytopenia (11/14), moderate to severe neutropenia (8/14), and/or normocytic anemia (8/14). There was evidence of immune-mediated destruction of peripheral blood cells. All treated dogs exhibited a significant reduction in marrow colony-forming capacity, irrespective of whether peripheral cytopenia was present, with 12/14 showing decreased CFU-GM and 14/14 decreased CFU-E activity. Within a week following cessation of dosing, all affected dogs achieved hematologic remission as defined by restoration of the peripheral blood counts. However, despite this apparent recovery, both CFU-E and CFU-GM activities of the bone marrow remained depressed for at least another 8 months. We conclude that in dogs prolonged administration of high doses of cefazedone induced a persistent deficit of CFU-E and CFU-GM progenitor cells. The clinical relevance of this, if any, remains to be established.

Reproducible cloning assays for in vitro growth of canine hematopoietic progenitor cells and their potential applications in investigative hematotoxicity.

A variety of in vitro cloning assays have been used for studying hematopoiesis in mice and human beings. However, these techniques have had limited use in dogs, a species used extensively as a model for hematopoietic research, particularly hematotoxicity. We have adopted cloning assays for in vitro growth of canine colony-forming unit-erythroid (CFU-E) and colony-forming unit-granulocyte/macrophage (CFU-GM) progenitor cells, using modified microplasma clot and soft agar culture systems, respectively. Marrow mononuclear cells separated by density-gradient centrifugation were added to the aforementioned culture systems. Erythroid colonies were stimulated with sheep plasma erythropoietin and incubated at 37 C in 5% CO2 for 2 days. The CFU-E colonies were fixed with 5% glutaraldehyde, stained with benzidine, counted, and expressed as a mean of 8 replicates. The CFU-GM colonies were stimulated with pooled serum from endotoxin-treated dogs and incubated for 8 days at 37 C in 10% CO2. Using an inverted microscope, the CFU-GM colonies were counted and expressed as a mean of 6 replicates. The number of colonies was proportional to the plated cell concentrations. The addition of 10% autologous serum to CFU-GM cultures increased the number of colonies by 80 to 100%, but markedly reduced the size and number of CFU-E colonies. The marrow cloning capacity among dogs of comparable age was similar, and little variation was noticed when bone marrow cells from the same dogs were cultured repeatedly over a period of 3 to 4 months. We concluded that these cloning assays are fast, reliable, and reproducible and that they allow quantitative determination of canine hematopoietic progenitor cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Cephalosporin-induced immune cytopenia in the dog: demonstration of erythrocyte-, neutrophil-, and platelet-associated IgG following treatment with cefazedone.

Cephalosporin treatment in man has been associated with a low incidence of hemolytic anemia, thrombocytopenia, and neutropenia; some cases have been shown to be immune-mediated. This triad of blood dyscrasias was also demonstrated in our laboratory in a series of toxicity studies in dogs of two cephalosporin compounds, cefonicid and cefazedone; these studies provided evidence for drug-associated immune hemolytic anemia, based on conventional laboratory tests. To further investigate possible immune mechanisms of the cephalosporin-induced cytopenias, we measured erythrocyte-associated, platelet-associated (PAIgG), and serum antineutrophil IgG over the course of cephalosporin treatment, using highly sensitive 125I-staphylococcal protein A (SPA) assays, as well as the direct antiglobulin test; we compared these findings with the hematologic changes. Intravenous treatment with high doses of cefazedone (540 mg/kg/day, increased to a maximum of 840 mg/kg/day for 4 months or until hematologic effects were evident) resulted in a high incidence of anemia (7/14), thrombocytopenia (11/14), and neutropenia (7/14). Of the affected dogs examined, 6/7 with anemia, 9/9 with thrombocytopenia, and 7/7 with neutropenia showed increased levels of the respective cell-associated antibody, compared with untreated controls. Unaffected dosed animals generally did not show these changes. In 3/3 dogs examined following remission of thrombocytopenia, PAIgG returned to levels comparable with controls; as one of these dogs suffered a relapse, increased PAIgG was again observed. Animals sacrificed during cytopenic episodes showed cytologic and histologic evidence of increased hemophagocytosis. We conclude that antibody-mediated blood cell destruction contributes to all three cephalosporin-induced cytopenias in the dog.

Cephalosporin-induced changes in the ultrastructure of canine bone marrow.

Fourteen healthy dogs were given 540 to 840 mg/kg of cefazedone (Refosporen) intravenously for up to 4 months or until peripheral blood cell count were depressed. Within 6 to 10 weeks treated dogs developed pancytopenia (5/14), thrombocytopenia (11/14), moderate to severe neutropenia (8/14), and/or normocytic anemia with erythroblastemia (8/14). Ultrastructural changes in bone marrow of severely cytopenic dogs included mitochondrial damage in hematopoietic and nonhematopoietic cells, thickening of endosteal bone lining layers, increased adventitial coverage of vascular sinuses, and an increased number of active macrophages. Swollen, ruptured mitochondria were in erythroid, granulocytic, and megakaryocytic cells, and, to a lesser extent, in macrophages, reticular endothelial, and bone lining cells. Maturation arrest was evident in both erythroid and granulocytic cell lines. There was also evidence of ineffective erythropoiesis and granulopoiesis. None of these changes were observed in bone marrow of controls, treated dogs that did not develop cytopenia, or dogs allowed to recover after cessation of dosing.

The hematologic effects of cefonicid and cefazedone in the dog: a potential model of cephalosporin hematotoxicity in man.

Cephalosporin antibiotics cause a variety of hematologic disturbances in man, the pathogeneses and hematopathology of which remain poorly characterized. There is a need for a well-defined animal model in which these blood dyscrasias can be studied. In four subacute toxicity studies, the intravenous administration of cefonicid or cefazedone to beagle dogs caused a dose-dependent incidence of anemia, neutropenia, and thrombocytopenia after 1-3 months of treatment. A nonregenerative anemia was the most compromising of the cytopenias and occurred in approximately 50% of dogs receiving 400-500 mg/kg cefonicid or 540-840 mg/kg cefazedone. All three cytopenias were completely reversible following cessation of treatment; the time required for recovery of the erythron (approximately 1 month) was considerably longer than that of the granulocytes and platelets (hours to a few days). Upon rechallenge with either cephalosporin, the hematologic syndrome was reproduced in most dogs tested; cefonicid (but not cefazedone)-treated dogs showed a substantially reduced induction period (15 +/- 5 days) compared to that of the first exposure to the drug (61 +/- 24 days). This observation, along with the rapid rate of decline in red cell mass parameters of affected dogs, suggests that a hemolytic component complicated the red cell production problem and that multiple toxicologic mechanisms contributed to the cytopenia. We conclude that the administration of high doses of cefonicid or cefazedone to dogs can induce hematotoxicity similar to the cephalosporin-induced blood dyscrasias described in man and thus provides a useful model for studying the mechanisms of these disorders.

The hematopathology of cefonicid- and cefazedone-induced blood dyscrasias in the dog.

Cephalosporin treatment in man has been associated with blood dyscrasias that include a time- and dose-related anemia, neutropenia, and thrombocytopenia, the hematopathology of which remains poorly characterized. A similar hematologic syndrome can be produced in dogs following daily intravenous injections of 540-840 mg/kg cefazedone or 400-500 mg/kg cefonicid for 1-3 months. Using this animal model, histologic and cytologic changes in blood, bone marrow, spleen, and liver were studied over the course of the cephalosporin-induced cytopenias. Peripheral blood cytologic observations included an absence, generally, of erythroid regenerative changes, increased numbers of macroplatelets, spherocytosis, erythroblastemia, and toxic neutrophil morphology. Interim and postmortem cytologic and histologic observations of bone marrow included hypoplastic and toxic changes, primarily in cytopenic dogs receiving high doses of cefonicid, and regenerative changes in hematopoietic tissue of affected cefazedone-treated animals. The latter included variable erythroid hyperplasia, increased megakaryocytes, and decreased marrow fat and was accompanied by evidence of extra-medullary hematopoiesis and increased hemosiderin and hemophagocytosis in liver and splenic tissue. The incidence and severity of these changes were dose-dependent, corresponded with the cytopenias observed peripherally, and, like the cytopenias, were fully reversible. These observations suggest that the hematologic syndrome associated with cephalosporin treatment in the dog has multiple toxicologic mechanisms, which include peripheral cytotoxic effects and bone marrow damage with depressed or ineffective hematopoiesis.

Bone lining cells and hematopoiesis: an electron microscopic study of canine bone marrow.

Hematopoietic bone marrow in the dog is enclosed by a nearly complete and rather complex layer of endosteum, consisting of a diverse group of cells collectively called bone lining cells (BLC). Cell types comprising BLC include osteoblasts and osteoclasts, and other cell types, among which are elongated, flat cells with a spindle-shaped nucleus, and small cytoplasmic vesicles. The composition and thickness of the layer of BLC varies along the perimeter of the marrow. The layer may be simple or stratified. Occasionally a zone of tightly packed regularly arranged collagenous fibers lies between the bone lining cells and bone. Hematopoiesis, particularly neutrophilic, often occurs in the bone marrow next to the BLC. Cytoplasmic processes of BLC occasionally extend into the hematopoietic spaces and stromal cells in the hematopoietic compartment may extend processes to the layer of BLC. Occasionally cells of the BLC are similar in appearance to stromal cells within the marrow. Our observations together with the experimental findings of others (that fibroblastic stromal cells contribute to the hematopoietic inductive microenvironment, that hematopoietic stem cells are concentrated subosteally, that cells responsible for regeneration of the marrow stroma are derived from the endosteal layer, and that high concentrations of hematopoietic colony-stimulating factors are produced there) indicate that the hematopoietic capacities of bone marrow may be regulated by BLC.

Effects of Escherichia coli endotoxin on leukocyte and platelet counts, fibrinogen concentrations, and blood clotting in colostrum-fed and colostrum-deficient neonatal calves.

Effects of a single IV injection of Escherichia coli endotoxin on hemogram and clotting function were compared in colostrum-fed and colostrum-deficient neonatal calves. Before endotoxin administration, the 2 groups of calves only differed in their prothrombin times. After endotoxin administration, there were significant differences (P less than 0.005) between colostrum-fed and colostrum-deficient calves in total leukocyte, segmented neutrophil, nonsegmented neutrophil, and lymphocyte (P less than 0.05) counts and partial thromboplastin time. Significant time dependent changes were observed in the aforementioned parameters and in platelet count and fibrinogen concentration. Seemingly, colostrum feeding improved the calf's ability to respond to endotoxin challenge exposure probably because of improved granulopoietic activity.

Changes in selected biochemical constituents of blood collected from horses participating in a 50-mile endurance ride.

The effects of strenuous exercise on serum electrolytes, blood metabolites, and serum enzymes were studied in a group of 13 horses participating in a 50-mile endurance ride. Blood samples were collected before, during, and at the end of the ride, as well as 1 hour and 16 hours after the completion of the ride. There were significant changes in these values when preride values were compared with those of samples taken at different sample-collection periods. Significant (P less than 0.001) decreases were observed in serum concentrations of chloride, potassium, and calcium. A significant increase in the serum concentration of phosphorous also was observed. The mean concentrations of chloride and calcium remained significantly reduced in the 16-hour postride sample. A small, but significant (P less than 0.05), increase in sodium and decrease in glucose concentrations were observed at the midride sample. Although midride samples did not reflect a significant change in blood pyruvate concentration, a significant (P less than 0.05) increase was observed at the end of the ride. Changes in the hematocrit and the total protein concentration were statistically significant at levels of P less than 0.001 and P less than 0.01, respectively. The horses showed highly significant increases in concentrations of free fatty acids, serum enzymes (P less than 0.001), and lactate (P less than 0.005) during the ride. The elevated values of creatine kinase and lactic dehydrogenase decreased, while the aspartate aminotransferase values remained increased long after completion of the ride. The lactate and free fatty acids, which rose significantly during the ride, decreased to approximately preride values by 16 hours after the ride was completed. We conclude that the changes in serum electrolytes, enzymes, and blood metabolites reflect the sweating, muscle damage, and increased dependency on anaerobic glycolysis and lipid mobilization that these horses experienced during the competitive endurance ride.